Local Derivative Pattern with Smart Thresholding: Local Composition Derivative Pattern for Palmprint Matching

Abstract

Palmprint recognition is a new biometrics system based on physiological characteristics of the palmprint, which includes rich, stable, and unique features such as lines, points, and texture. Texture is one of the most important features extracted from low resolution images. In this paper, a new local descriptor, Local Composition Derivative Pattern (LCDP) is proposed to extract smartly stronger and more distinguishing texture features from palmprint images by composition of both radial and directional derivative information among local neighbors using a threshold function with an adaptive threshold value which result from local directional derivative information. The distribution of the LCDP is modeled by local spatial histogram and histogram intersection function is used to measure the similarity between spatial histograms of two different palm print images. Then, nearest neighbor classifier is used to classify them. Experiments on the Hong Kong Polytechnic University (PolyU) 2D_3D_palmprint database demonstrate the effectiveness of the LCDP in palmprint recognition versus well-known local pattern descriptors.